The most award winning
healthcare information source.
TRUSTED FOR FOUR DECADES.
Source: Christenson J, et al. Early discharge of patients with presumed opioid overdose: Development of a clinical prediction rule. Acad Emerg Med 2000;7:1110-1118.
Emergency medicine and toxicology texts commonly recommend observing patients after opioid overdose for four to 24 hours after the last dose of an opioid antagonist. This suggestion is based on concerns that longer-acting opioids may cause recurrent respiratory depression or delayed onset pulmonary edema. This study from St. Paul’s Hospital in Vancouver, Canada, sought to develop a clinical prediction rule to identify patients who may be safely discharged only one hour after the last dose of the opioid antagonist, naloxone.
This prospective cohort study included 573 patients who received naloxone either 0.4 mg intravenously (IV) or 0.8 mg subcutaneously (SQ) during pre-hospital care or in the emergency department (ED). Exclusion criteria included death within one hour of naloxone administration, not having a name documented on the chart, and refusal to consent to follow-up. The investigators recorded the time of naloxone administration, and a formal assessment was planned for one hour later. If naloxone was re-administered prior to the completion of that hour, the clock was restarted and assessment again was planned for one hour later.
Telephone follow-up was attempted for patients who provided phone numbers. Subjects were asked whether they returned to the hospital for any reason within 24 hours of their one-hour assessment. If direct contact could not be made, indirect contacts (friends, family) were attempted to confirm patient status 24 hours post-discharge. A list of subjects who could not be contacted was matched with medical record databases of the six other hospitals in greater Vancouver, Department of Vital Statistics records, and the coroner’s office records. For any return hospital visit within 24 hours, the reason for the visit and any adverse events were abstracted from the medical record.
Patients were classified into two groups: those with adverse events within 24 hours and those without. The investigators used classification and regression tree (CART) methodology to develop a decision rule to predict safe discharge. The rule predicted that patients with presumed opioid overdose can be discharged safely one hour after naloxone administration if they have: 1) the ability to mobilize as usual; 2) an oxygen saturation on room air greater than 92%; 3) a respiratory rate between 10-20 breaths/min; 4) a temperature between 35-37.5°C; 5) a heart rate between 50-100 beats/min; and 6) a GCS of 15. The prediction rule had a sensitivity of 99% and a specificity of 40% for predicting adverse outcomes within the first 24 hours.
This well-done, prospective study has made great strides toward identifying patients who may be safely discharged one hour after the administration of naloxone for opioid overdose. However, it does have several limitations, including direct phone follow-up with only 20% of subjects. However, commendable efforts were made toward contacting patients’ friends and relatives and identifying outside hospital visits and deaths. Loss of subjects due to migration is unlikely, as the authors included visits to other Vancouver hospitals and the follow-up period was only 24 hours.
In addition, the authors do not specify the number and type of presumed opioid overdoses other than heroin. Nearly 86% of the study subjects admitted to heroin use prior to their ED visit, and others may have used heroin but denied it during the ED interview. This overwhelming majority of heroin overdoses leads me to believe that this prediction rule may not be applicable to patients who overdose on opioids other than heroin (e.g., orally ingested or longer-acting agents).
Lastly, it is important to note that this project was the development phase of this particular clinical prediction rule and may not be generalizable to other settings. The drug overdose profiles (e.g., mixing of opioids with other drugs of abuse) can differ significantly from region to region, as can practices regarding naloxone administration. In this study, 88% of patients received 0.8 mg naloxone SQ, which may have a longer duration of action than IV naloxone. This may differ from the standard route and dose of naloxone in other hospitals and EMS systems.